Amazon Cloud Technology Build On 2022 - AIot Season 2 IoT Special Experiment Experience

大家好

Hello,各位好,Very glad to second participated in theAWS&CSDN举办的 Build On 活动,This period of activity is the theme of the Internet of things

Build On是 什么

Amazon's cloud technology developersBuild OnIs planned by the amazon team、The developer community together to use series of activities.It is at the core reality technology application and demand scenarios,Combining with the current key technology
And amazon's cloud technology at the forefront of technology scheme of,面向开发人员、IT技术人员、Necessary cloud or technical decision makers in the field of course.
2022年亚马逊云科技Build OnSeries of activities will revolve around data、软件、架构、Operations in cutting-edge technologies in the field of core technology areas,Aims to provide professional and technical direction of experiment、Ta refers to
导、Experts answer questions, etc,Help developers to understand the related classic technology framework and classic example of best practice in the field of,And ultimately through the elaborate design experiment process environment,Led by technical experts hold developers set personally
计、部署和操作.Topics will cover the cloud intoiBasic and professional service application,如机器学习、loT技术、Serverless、 基础设施等,Cover from startups to mature enterprise full scene full of life
Cycle of business practical cases,Whether you are new to developers to the cloud,Or development experienced expert,You will be fromBuild OnActivities material gain.

本次Build On主题介绍

在居家安防监控领域,基于实时视频的移动检测,发现监控环境中人、宠物、包裹等的出现,并且能实时地将检测结果通知给身处任何地方的用户是其重要的应用场景之一.但在这一场景的技术实现中面临如下的挑战：一是基于摄像头的视频检测通知,存在大量由于风、雨、移动的车等并非用户关注的事件误报,严重影响用户的使用体验. 二是实现这一方案涉及的技术领域与复杂度很高,如设备端事件检测和触发、视频编解码处理、视频存储、机器视觉等,需要团队具备较强的技术和专业能力.This experiment will be to minimize the prototype,Reflected by the Raspberry Pi Zero 2 W Add cameras as security devices,并使用 Amazon KVS 和 Amazon Rekognition Streaming Video Events How to solve the above challenge,Real-time intelligent visual identification.
活动链接：https://marketing.csdn.net/p/dba35524bec59472d5b2e1e7b48b7403
实验手册：https://aws.amazon.com/cn/getting-started/hands-on/intelligent-visual-recognition-with-kvs-rekognition/

To participate in this activity you can learn what

• The cameraIoTEquipment integrated development skills
• The video streaming data unified management and scheduling between cloud open 发技能
• Study on the cloud video processing andAlVisual identification test method
• Complete cloud based on identification test results trigger device operation completeAloTA closed-loop best practice experience

This experiment usedAWS服务

• Amazon IAM
• Amazon S3
• Amazon KVS
• Amazon Cloud9
• Amazon Rekognition
• Amazon SNS
  

实验部分

1.线上报名(2022.08.06)

With no opportunity to participate in the activities of offline buddy（比如博主自己）You can join online,The active side here very close consideration to this,Online users can directly on their own account to do experiments or contact activities little helper applications,There is also another field is in 2022.08.06There is a line experiment,We are interested in can contact little helper applications
CSDN小助手(WeChat):CSDNCS010

2.实验过程

Experimental explanation set up video

视频链接地址:https://www.bilibili.com/video/BV1dS4y1t7SK/

2.1、流程

2.1.1、注意事项

1、Maintain consistent service areas

Please participate in the experiment the students must pay attention to keep all the service's operations in the same area,For example, the experiment mentioned in the manual 爱尔兰(eu-west-1) 或者 I in the video demo for you Northern Virginia(us-east-1)

2、使用IAM用户操作

Especially need to pay attention to must useIAM用户进行操作,请勿使用ROOT用户(Is the email account)进行操作,May lead to can't receive email

3、服务ARN

Please especially pay attention to create the serviceARN,将其记录下来,Or as a demonstration of my video,Each service open a TAB,以便后面使用

2.1.2、错误解决

1、如在cloud9Update the source and the installation package encountered the following error

E: Could not get lock /var/lib/dpkg/lock-frontend - open (11: Resource temporarily unavailable)
E: Unable to acquire the dpkg frontend lock (/var/lib/dpkg/lock-frontend), is another process using it?

So please run,然后将 apt 更换为 apt-get

sudo rm /var/lib/dpkg/lock-frontend
sudo rm /var/lib/dpkg/lock

2、 AWS::S3::PermanentRedirect

Please confirm whether your buckets and other services in the same area

3、json格式错误

Please note that in the create and startRekognitionVideo streaming eventsjson中是否包含错误,建议直接在cloud9To write and modify

4.创建create-stream-processor错误

First check whether you have created before,用如下命令查看

aws rekognition list-stream-processors

如果有,So please delete,再创建.

#删除命令
aws rekognition delete-stream-processor --name 已创建的processor名字

If you want to change the name to create,那么直接修改json中的 processor的名字即可

2.2、所使用的命令

2.2.1、rekognition 部分

#创建
aws rekognition create-stream-processor -region Your area --cli-input-json 你的json文件
#描述
aws rekognition describe-stream-processor --name processor名称 --region 区域
#列出
aws rekognition list-stream-processors
#启动
aws rekognition start-stream-processor --region Your area --cli-input-json 你的json文件
#删除
aws rekognition delete-stream-processor --name processor名称 --region 区域

2.2.2、S3部分

#列出文件
aws s3 ls 存储桶名称 --recursive

#Remove all in bucket object
aws s3 rm s3://存储桶名称 --recursive

#Removal of bucket
aws s3 rb s3://存储桶名称

2.2.3、SNS部分

#清除 topic
aws sns delete-topic --topic-arn <您的topic arn>

#清除订阅
aws sns unsubscribe --subscription-arn <您的subscripiton arn>

2.2.4、KVS

#清除 Kinesis video stream
aws kinesisvideo delete-stream --stream-arn <您的stream arn>

3、实验结果

3.1、Task check

3.1.1、SNSEmail to subscribe to check

3.1.2、SNSNotification to the email address

3.1.3、S3File is written to see

4、总结

The second time I participate inAWS的Build On活动,也是Build On的第二季,It's a pity that in the experiment because there is no development board,Can't do in front of the tree blackberry send link of experiment,The overall process down is really very simple,就是在Cloud9It cost for a long time,You can also like I demonstrated in the video,合理的利用时间,The whole experiment took them even less time.In this experiment have some harvest,例如那个S3的错误,以及启动 rekognition-stream-processor 的jsonFile defined the timestamp of the,That is to operation according to their own video,Can also listen to me in the video description.So this blog is over here,希望您在2022.08.06Before reading this blog,And well to attend to 2022年8月6Day online experiment,I wish you gain full

感谢阅读